Belder, D.J., Pierson, J.C., Ikin, K., Blanchard, W., Westgate, M.J., Crane, M., Lindenmayer, D.B
Total Page:16
File Type:pdf, Size:1020Kb
Belder, D.J., Pierson, J.C., Ikin, K., Blanchard, W., Westgate, M.J., Crane, M., Lindenmayer, D.B. (2019) Is bigger always better? Influence of patch attributes on breeding activity of birds in box-gum grassy woodland restoration plantings. Biological Conservation, Vol. 236, pp. 134-152. DOI: https://doi.org/10.1016/j.biocon.2019.05.015 © 2019. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ 1 Is bigger always better? Influence of patch attributes on breeding activity of 2 birds in box-gum grassy woodland restoration plantings 3 *Donna J. BelderAB, Jennifer C. PiersonAC, Karen IkinA, Wade BlanchardA, Martin J. 4 WestgateA, Mason CraneAD, David B. LindenmayerABD 5 A Fenner School of Environment and Society, The Australian National University, Canberra, ACT, 6 Australia 2601. 7 B National Environmental Science Program Threatened Species Recovery Hub, The Australian 8 National University, Canberra, ACT, Australia 2601. 9 C ACT Parks and Conservation, Environment, Planning and Sustainable Development Directorate, 10 ACT Government, Canberra, ACT, Australia 2602. 11 D Sustainable Farms, The Australian National University, Canberra, ACT, Australia 2601. 12 Abstract 13 Restoration plantings are an increasingly common management technique to address habitat loss in 14 agricultural landscapes. Native fauna, including birds, may readily occupy planted areas of 15 vegetation. However, unless restoration plantings support breeding populations, their effectiveness 16 as a conservation strategy may be limited. We assessed breeding activity of birds in box-gum grassy 17 woodland restoration plantings in the South-west Slopes bioregion of New South Wales, Australia. 18 We compared breeding activity in plantings of different size (small and large) and shape (linear and 19 block-shaped) to breeding activity in a set of remnant woodland sites. Contrary to expectations, we 20 found that bird breeding activity was greatest per hectare in small patches. We also found a negative 21 effect of planting age, with younger plantings supporting more breeding activity per hectare. We 22 found no effect of patch type or shape on breeding activity, and that species’ relative abundance 23 was not predictive of their degree of breeding activity. Our results highlight the value of small 24 habitat patches in fragmented agricultural landscapes, and indicate that restoration plantings are as 25 valuable as remnant woodland patches for supporting bird breeding activity. We demonstrate the 1 26 importance of breeding studies for assessing the conservation value of restoration plantings and 27 other habitat patches for avifauna. 28 29 Keywords: Woodland birds, breeding success, SLOSS, restoration, fragmentation, agricultural 30 landscapes 31 1 32 1. Introduction 33 Habitat loss due to land conversion for agriculture is a significant issue globally, with numerous 34 effects on biodiversity and ecosystem processes (Maxwell et al., 2016; Tscharntke et al., 2012). 35 Land clearing is increasing worldwide, particularly in productive agricultural regions (Evans, 2016; 36 Tilman et al., 2017). The extensive removal of native vegetation creates a highly fragmented 37 landscape in which patches of native vegetation exist primarily as small, isolated remnants. 38 Restoration plantings in agricultural landscapes are increasingly implemented to address habitat loss 39 and conserve threatened and declining native fauna, with hundreds of millions of hectares of 40 vegetation being replanted around the world at a cost of billions of dollars (Crouzeilles et al., 2016). 41 To ensure cost-effectiveness and ecological integrity, it is important to quantify the effects of 42 revegetation on biodiversity and assess whether conservation goals are being met, particularly in the 43 long term (Barral et al., 2015; Ruiz-Jaen and Aide, 2005). 44 A core assumption of restoration success is that revegetated patches provide high-quality habitat for 45 the species they are intended to help conserve (Ikin et al., 2016; Ruiz-Jaen and Aide, 2005). In 46 Australia, bird communities that inhabit box-gum grassy woodlands are threatened by ongoing 47 habitat loss and degradation (Rayner et al., 2014), and are a frequent target of restoration efforts 48 (Freudenberger, 2001; Lindenmayer et al., 2013; Smith, 2008). There is evidence suggesting that 49 many bird species will readily occupy restoration plantings, in some cases preferentially inhabiting 50 plantings over remnant woodland patches or other sites (Barrett et al., 2008; Cunningham et al., 51 2008; Lindenmayer et al., 2016), but how much do we know about the capacity of restoration 52 plantings to support breeding populations of these species? The majority of studies examining avian 53 responses to restoration plantings have used measures such as species richness, diversity, and 54 relative abundance to make inferences about occupancy trends and habitat quality (Belder et al., 55 2018). However, focusing on occurrence patterns provides a limited picture of how birds are using a 56 site (Chalfoun and Martin, 2007). It is therefore important to quantify whether indicators of long- 57 term persistence, such as breeding activity, follow the same trends. 1 58 1.1. Research objectives 59 The underlying aim of this study was to assess whether birds are able to breed successfully in box- 60 gum grassy woodland restoration plantings. Breeding success can be measured in several ways, 61 with nest success and daily nest survival being commonly-used metrics (Stephens et al., 2004). 62 However, searching for, and monitoring, nests requires considerable time and effort. An alternative, 63 and perhaps more accessible, approach is to use indicators of breeding activity as a proxy for 64 breeding success. For example, a scoring system developed by Mac Nally (2007) ranks 65 observations of breeding behaviour according to how strongly they indicate breeding success (Table 66 1), providing a quantitative measure of the extent to which a given site supports successful breeding 67 (Bennett et al., 2015; Mac Nally et al., 2010; Selwood et al., 2009). A method such as this provides 68 a basis from which to commence the transition from traditional occupancy and abundance surveys 69 to a more population-oriented approach to monitoring avian responses to restoration plantings. 70 Importantly, it also facilitates the collection of breeding data on species of conservation concern, 71 whose nests may be difficult to find in adequate numbers. 72 We sought to investigate bird breeding activity in the context of habitat restoration in a fragmented 73 agricultural landscape. Specifically, we posed the following three questions: 74 Question 1. How does bird breeding activity in restoration plantings compare to breeding activity 75 in remnant woodland patches? 76 We compared breeding activity in restoration plantings, similar-sized woodland remnants, and 77 larger, more intact woodland remnants. In addition to investigating the entire bird assemblage, we 78 assessed breeding activity for species of conservation concern, and cup-nesters vs. dome nesters 79 (Appendix B). Remnant patches are generally considered to be high-value habitat within 80 fragmented agricultural landscapes (Cunningham et al., 2014), and hence we predicted remnant 81 sites would support more breeding activity than restoration plantings. We predicted that breeding 82 activity would be highest in larger woodland remnants than in smaller, more isolated remnants and 1 83 restoration plantings. We made this prediction because comparative studies have shown that species 84 richness and abundance is typically highest in large, intact remnants (Hadley et al., 2018; Helzer 85 and Jelinski, 1999; Martin et al., 2004; Munro et al., 2011). Many species of conservation concern 86 are more closely associated with remnants than plantings (Kinross, 2004), so we also expected to 87 observe more breeding activity from these species in remnants than in plantings. 88 Question 2. How do patch attributes affect breeding activity in plantings and remnant woodland 89 patches? 90 We examined breeding activity in sites of varying size (small and large) and shape (linear and 91 block-shaped). A key finding from pattern-based studies of bird distribution and abundance in 92 fragmented landscapes is that larger patches support more species (Kavanagh et al., 2007; Shanahan 93 et al., 2011; Watson et al., 2003). This is consistent with the resource concentration hypothesis, 94 which posits that there are more resources and thus more individuals and greater species diversity in 95 larger patches (Connor et al., 2007; Root, 1973). Previous species-specific studies have also found 96 that avian reproductive success is positively correlated with patch size (Herkert et al., 2003; Hoover 97 et al., 1995; Luck, 2003; Zanette et al., 2000). We therefore postulated that breeding activity would 98 increase with patch size in parallel with bird species richness and abundance. Similarly, increasing 99 patch linearity is typically associated with lower species richness and abundance (Kinross, 2004; 100 Lindenmayer et al., 2018a, 2007). As such, we predicted more evidence of successful breeding in 101 block-shaped than in linear patches. 102 We predicted a stronger negative response to decreasing patch size and increasing linearity for cup- 103 nesters compared with dome-nesters. This was because edge-effects of predation are stronger in 104 smaller and more linear sites